arXiv:0901.1099v1 [q-fin.PR] 8 Jan 2009

Impact of volatilities and correlation Damiano Brigo Fitch Solutions and Dept. of Mathematics Imperial College

Kyriakos Chourdakis Fitch Solutions and CCFEA Univ. of Essex

Imane Bakkar Fitch Solutions 101 Finsbury Pavement London

Abstract It is commonly accepted that Commodities futures and forward prices, in principle, agree under some simplifying assumptions. One of the most relevant assumptions is the absence of counterparty risk. Indeed, due to margining, futures have practically no counterparty risk. Forwards, instead, may bear the full risk of default for the counterparty when traded with brokers or outside clearing houses, or when embedded in other contracts such as swaps. In this paper we focus on energy commodities and on Oil in particular. We use a hybrid commodities-credit model to asses impact of counterparty risk in pricing formulas, both in the gross effect of default probabilities and on the subtler effects of credit spread volatility, commodities volatility and credit-commodities correlation. We illustrate our general approach with a case study based on an oil swap, showing that an accurate valuation of counterparty risk depends on volatilities and correlation and cannot be accounted for precisely through a pre-defined multiplier.

AMS Classification Codes: 60H10, 60J60, 60J75, 62H20, 91B70 JEL Classification Codes: C15, C63, C65, G12, G13 Keywords: Counterparty Risk, Credit Valuation adjustment, Commodities, Swaps, Oil models, Convenience Yield models, Stochastic Intensity models. Fitch Solutions, 101 Finsbury Pavement, EC2A 1RS London E-mail: [email protected] June 24, 2008. Available also at www.damianobrigo.it and at SSRN.com

Brigo, D., Bakkar, I., and Chourdakis, K., Counterparty risk for Energy-Commodities swaps

2

1 Introduction In this paper we consider counterparty risk for commodities payoffs in presence of correlation between the default event and the underlying commodity, while taking into account volatilities both for credit and commodities. We focus on Oil but much of our reasoning can be adapted to other commodities with similar characteristics (storability, liquidity, and similar seasonality). Past work on pricing counterparty risk for different asset classes is in Sorensen and Bollier (1994), Brigo and Masetti (2006) and Brigo and Pallavicini (2007, 2008) for interest rate swaps and exotics underlyings. Leung and Kwok (2005) and Brigo and Chourdakis (2008) worked on counterparty risk for credit (CDS) underlyings. Here we analyze in detail counterparty-risky (or default-risky) Oil forward and swaps contracts. In general the reason to introduce counterparty risk when evaluating a contract is linked to the fact that many financial contracts are traded over the counter, so that the credit quality of the counterparty can be relevant. This is particularly appropriated when thinking of the different defaults experienced by some important companies during the last years, especially in the energy sector. Earlier works in counterparty risk for commodities include for example Cannabaro, Picoult and Wilde (2005), who analyze this notion more from a capital adequacy/ risk management point of view. In particular, their approach is not dynamical and does not consider explicitly credit spread volatility and especially correlation between the underlying commodity and credit spread. In our approach wrong way risk is modeled through said correlation. Mostly, however, the difference is in the purpose. We are valuing counterparty risk more from a pricing than a risk management perspective, resorting to a fully arbitrage free and fine-tuned risk neutral approach. This is why all our processes are calibrated to liquid market information both on forward curves and volatilities. Correlations are harder to estimate but we analyze their impact by letting them range across a set of possible values. A limitation of our approach is that we refer to a single counterparty. In general we are looking at the problem from the viewpoint of a safe (default-free) institution entering a financial contract with another counterparty having a positive probability of defaulting before the final maturity. We formalize the general and reasonable fact that the value of a generic claim subject to counterparty risk is always smaller than the value of a similar claim having a null default probability, expressing the discrepancy in precise quantitative terms. We consider Credit Default Swaps for the counterparty as liquid sources of market default probabilities. Different models can be used to calibrate CDS data and obtain default probabilities: here we resort to Brigo and Alfonsi (2005) stochastic intensity model, whose jump extension with analytical formulas for CDS options is illustrated in Brigo and El-Bachir (2008). As a model for oil we adopt a two factor model shaping both the short term deviation in prices and the equilibrium price level, as in Smith and Schwartz (2000). This model can be shown to be equivalent to a more classical convenience yield model like in Gibson and Schwartz (1990), and a stochastic volatility extension of a similar approach is considered in Geman (2000). What is modeled is the oil spot price, under the implicit assumption that such a spot price process exists. This is not true for electricity, for example, and even for markets like crude oil where spot prices are quoted daily, the exact meaning of the spot is difficult to single out. Nonetheless, we assume, along with most of the industry and with Carmona and Ludkowski (2004), that there is a traded spot asset. In the paper we find that counterparty risk has a relevant impact on the products prices and that, in turn, correlation between oil and credit spreads of the counterparty has a relevant

Brigo, D., Bakkar, I., and Chourdakis, K., Counterparty risk for Energy-Commodities swaps

3

impact on the adjustment due to counterparty risk. Similarly, oil and credit spread volatilities have reasonable impacts on the adjustment. The impact patterns do not involve the peculiar behaviour one observes in the case of credit underlyings, observed in Brigo and Chourdakis (2008). Nonetheless, the impact is quantitatively relevant, and we illustrate this with a case study based on an oil swap. The paper is organized as follows: Section 2 lays down the general framework for the valuation of counterparty risk. In section 3 we present the CIR++ specification which serves as the credit model, and in section 4 we outline the two-factor Smith and Schwartz commodity model. Sections 5 and 6 illustrate the counterparty adjustments for forwards and swaps respectively. An example, based on a swap contract with between a bank and an airline company is presented in section 7.

2 General valuation of counterparty risk We denote by τ the default time of the counterparty and we assume the investor who is considering a transaction with the counterparty to be default-free. We place ourselves in a probability space (Ω, G, Gt , Q). The filtration (Gt )t models the flow of information of the whole market, including credit and defaults. Q is the risk neutral measure. This space is endowed also with a rightcontinuous and complete sub-filtration Ft representing all the observable market quantities but the default event (hence Ft ⊆ Gt := Ft ∨ Ht where Ht = σ({τ 6 u} : u 6 t) is the right-continuous filtration generated by the default event). We set Et (·) := E(·|Gt ), the risk neutral expectation leading to prices. Let us call T the final maturity of the payoff we need to evaluate. If τ > T there is no default of the counterparty during the life of the product and the counterparty has no problems in repaying the investors. On the contrary, if τ 6 T the counterparty cannot fulfill its obligations and the following happens. At τ the Net Present Value (NPV) of the residual payoff until maturity is computed: If this NPV is negative (respectively positive) for the investor (defaulted counterparty), it is completely paid (received) by the investor (counterparty) itself. If the NPV is positive (negative) for the investor (counterparty), only a recovery fraction REC of the NPV is exchanged. Let us call ΠD (t, T ) (sometimes abbreviated into ΠD (t)) the discounted payoff of a generic claim at t under counterparty risk. This is the sum of all cash flows from t to T , each discounted back at t, and under counterparty risk. This is a stochastic payoff, whose price would be given by risk neutral expectation. We denote by Π(t, T ) the analogous quantity when counterparty risk is absent, or when the counterparty is default free. All payoffs are seen from the point of view of the “investor” (i.e. the company facing counterparty risk). Then we have NPV(τ ) = Eτ {Π(τ, T )} and ΠD (t)

=

1{τ >T } Π(t, T ) + i h + + 1{t t}, the price of our payoff under counterparty risk is Et {ΠD (t)} = Et {Π(t)}−

+

LGD Et {1{t t)model = E(e−λ(t) ) = E exp (−Ψ(t, β) − Y (t))

(3.5)

Brigo, D., Bakkar, I., and Chourdakis, K., Counterparty risk for Energy-Commodities swaps

6

we just need to make sure E exp (−Ψ(t, β) − Y (t)) = Q(τ > t)market from which Ψ(t, β) = ln

E(e−Y (t) ) Q(τ > t)market

= ln

P CIR (0, t, y(0); β) Q(τ > t)market

(3.6)

where we choose the parameters β in order to have a positive function ψ (i.e. an increasing Ψ) and P CIR is the closed form expression for bond prices in the time homogeneous CIR model with initial condition y0 and parameters β (see for example Brigo and Mercurio (2001, 2006)). Thus, if ψ is selected according to this last formula, as we will assume from now on, the model is easily and automatically calibrated to the market survival probabilities for the counterparty (possibly stripped from CDS data). Once we have done this and calibrated CDS data through ψ(·, β), we are left with the parameters β, which can be used to calibrate further products. However, this will be interesting when single name option data on the credit derivatives market will become more liquid. Currently the bid-ask spreads for single name CDS options are large and suggest to either consider these quotes with caution, or to try and deduce volatility parameters from more liquid index options. At the moment we content ourselves of calibrating only CDS’s. To help specifying β without further data we set some values of the parameters implying possibly reasonable values for the implied volatility of hypothetical CDS options on the counterparty.

4 Commodity model We consider crude oil as a first important case. Suppose we have a airline company that buys a forward contract on oil from a bank with a very high credit quality, so that we assume the bank to be default-free. The bank wants to charge counterparty risk to the airline in defining the forward price, as there is no collateral posted and no margining is occurring. As a model for oil we adopt a two factor model shaping both the short term deviation in prices and the equilibrium price level, as in Smith and Schwartz (2000). This model can be shown to be equivalent to a more classical convenience yield model like in Gibson and Schwartz (1990), and a stochastic volatility extension of a similar approach is considered in Geman (2000). What is modeled is the oil spot price, under the implicit assumption that such a spot price process exists. This is not true for electricity, for example, and even for markets like crude oil where spot prices are quoted daily, the exact meaning of the spot is difficult to single out. Nonetheless, we assume, along with most of the industry, that there is a traded spot asset. If we denote by St the oil spot price at time t, the log-price process is written as ln(St ) = x(t) + L(t) + ϕ(t), where, under the risk neutral measure, dx(t)

= −kx x(t)dt + σx dZx ,

(4.1)

dL(t)

= µL dt + σL dZL ,

(4.2)

dZx dZL = ρx,L dt,

Brigo, D., Bakkar, I., and Chourdakis, K., Counterparty risk for Energy-Commodities swaps

7

and ϕ is a deterministic shift we will use to calibrate quoted futures prices. The process x represents the short term deviation, whereas L represents the backbone of the equilibrium price level in the long run. For applications it can be important to derive the transition density of the spot commodity in this model. For the two factors we have a joint Gaussian transition, #! " σ2 x x(t) x(s) exp(−kx (t − s)) (1 − exp(−2k (t − s))) Cov (s, t) x x,L ∼N , 2kx 2 L(t) x(s),L(s) L(s) + µL (t − s) σL (t − s) σx σL Covx,L (s, t) = ρx,L (1 − exp(−kx (t − s))). kx This can be used for exact simulation between times s and u. As we know that the sum of two jointly Gaussian random variables is Gaussian, we have ln(S(t)) = x(t) + L(t) + ϕ(t)|x(s),L(s) ∼ N (m(t, s), V (s, t)) m(t, s) = x(s) exp(−kx (t − s)) + L(s) + µL (t − s) + ϕ(t), σ2 2 V (t, s) = x (1 − exp(−2kx (t − s))) + σL (t − s) + 2Covx,L (s, t) 2kx from which, in particular, we see that E[S(t)|x(s), L(s)] = exp(x(s) exp(−kx (t − s)) + L(s) + µL (t − s) + ϕ(t) + V (s, t)/2) Hence we can compute the forward price E[S(T )|x(t), L(t)] at time t of the commodity at maturity T when counterparty risk is negligible and under deterministic interest rates, as F (t, T ) = exp(x(t) exp(−kx (T − t)) + L(t) + µL (T − t) + ϕ(T ) + V (T, t)/2)

(4.3)

In particular, given the forward curve T 7→ F M (0, T ) from the market, the expression for the shift ϕM (T ) that makes the model consistent with said curve is ϕM (T ) = ln(F M (0, T )) − x0 exp(−kx T ) − L0 − µL T − V (T, t)/2. The short term/equilibrium price model (x, L), when ϕ = 0, is equivalent to the more classical Gibson and Schwartz (1990) model, formulated as

d ln(St ) dq(t)

=

(r(t) − q(t) − σS2 /2)dt + σS dZS ,

= kq (α − q(t))dt + σq dZq ,

dZS dZq = ρq,S dt,

(4.4)

Brigo, D., Bakkar, I., and Chourdakis, K., Counterparty risk for Energy-Commodities swaps

8

the relationships being

kx

=

1 (q(t) − α) kq 1 ln(St ) − (q(t) − α) kq kq

µL

=

r − α − σS2 /2

σx

=

σq /kq

2 σL

=

σS2 + σq2 /kq2 − 2ρq,S σS σq /kq

dZx

=

dZL

=

ρx,L

=

dZq q σq σS dZS − dZq / σS2 + σq2 /kq2 − 2ρq,S σS σq /kq kq q σq σS ρq,s − / σS2 + σq2 /kq2 − 2ρq,S σS σq /kq kq

x(t)

=

L(t)

=

5 Forward vs Future prices and counterparty risk Consider now a forward contract. The propotypical forward contract agrees on the following. Let t be the valuation time. At the future time T a party agrees to buy from a second party a commodity at the price K fixed today. This is expressed by saying that the first party has entered a payer forward rate agreement. The second party has agreed to enter a receiver forward rate agreement. The value of this contract to the first and second party respectively, at maturity, will be ST − K, K − ST i.e. the actual price of the commodity at maturity minus the pre-agreed price in the payer case, and the opposite of this in the receiver case. Let us focus on the payer case. When this is discounted back at t with deterministic interest rates, and risk neutral expectation is taken, this leads to the price being given by Et [D(t, T )(ST − K)] = D(t, T )(Et [ST ] − K) = D(t, T )(F (t, T ) − K).

(5.1)

Note that the forward price is exactly the value of the pre-agreed rate K that sets the contract price to zero, i.e. K = F (t, T ). Let us maintain a general K in the forward contract under examination. In the oil model above, the forward contract price is given by plugging Formula (4.3) into (5.1). Let us denote by Fwdp(t, T ; K) such price (“p” is for payer), Fwdp(t, T ; K) = D(t, T ) (exp(x(t) exp(−kx (T − t)) + L(t) + µL (T − t) + ϕ(T ) + V (T, t)/2) − K) whereas the opposite of this quantity is denoted by Fwdr(t, T ; K). We may apply our counterparty risk framework to the forward contract, where now Π(t, T ) = D(t, T )(ST − K), and N P V (t) =Fwdp(t, T ; K). We obtain as price of the payer forward under

Brigo, D., Bakkar, I., and Chourdakis, K., Counterparty risk for Energy-Commodities swaps

9

counterparty risk from Equation (2.2). We obtain +

FwdpD (t, T ; K) = Fwdp(t, T ; K)− LGD Et {1{t